Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    words in a foreign language, possibly Russian or Japanese

    oai_token-act-pair · gpt-3.5-turbo

    special characters or symbols indicating formatting or encoding issues

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ierrez
    -0.87
     compr
    -0.78
    raints
    -0.77
    milo
    -0.74
     discont
    -0.74
    psey
    -0.72
     stake
    -0.70
     vested
    -0.70
    agents
    -0.69
     promoters
    -0.67
    POSITIVE LOGITS
    ħ
    1.41
    Į
    1.12
    ĩ
    1.09
    Ĩ
    1.04
    į
    1.03
    °
    0.98
    İ
    0.97
    Ī
    0.97
    Ļ
    0.96
    ¾
    0.95
    Act Density 0.005%

    No Known Activations